|
|
K-Nearest Neighbour Search Methods
Cigánik, Marek ; Bartík, Vladimír (referee) ; Burgetová, Ivana (advisor)
The thesis describes the basic concept of the K-nearest neighbors algorithm and its connection with the human concept of object similarity. Concepts and key ideas such as the distance function or the curse of dimensionality are elaborated. The work includes a detailed description of the methods KD-Tree, Spherical Tree, Locality-Sensitive Hashing, Random Projection Tree and families of algorithms based on the nearest neighbor graph. An explanation of the idea with visualizations, pseudocodes and asymptotic complexities is provided for each method. The methods were subjected to experiments and both basic and more advanced metrics were measured and appropriate use cases for individual methods were evaluated.
|
|
|
Data Mining Techniques
Kubincová, Monika ; Bartík, Vladimír (referee) ; Burgetová, Ivana (advisor)
The Bachelor's thesis deals with the processing and analysis of data from a commercial company, aiming to create an analytical tool for regular knowledge extraction from data that assists the company with important strategic decisions. The theoretical part of the thesis describes various methods of data mining and data processing, with a significant focus on the clustering method. The thesis further describes the available datasets that were used for the analysis and implementation of the proposed tasks. The final part of the concludes results of the analysis and its future usability including suggestions for improvement.
|
|
|
Time Series Analysis
Budai, Samuel ; Bartík, Vladimír (referee) ; Burgetová, Ivana (advisor)
This thesis deals with the issue of time series analysis and its use in the detection of anomalies in industrial networks. AR-X, ARIMA, SARIMA, Random Forest, Facebook Prophet and XGB Boost algorithms were used in the solution to create prediction models. In addition, the work includes the implementation of an algorithm for detecting anomalies from prediction models as well as solving the problem of high seasonal period in the case of the SARIMA algorithm. Through the conducted research, it was found that with the use of selected algorithms, it is possible to predict industrial traffic for the purpose of detection, within which up to 90% of attacks were detected. The work also provides a solution to a high seasonal period using partial time series. These results allow the experimental integration of prediction-based detection into real industrial networks.
|
|
|
Identification of Mobile Applications in Encrypted Traffic
Snášel, Daniel ; Burgetová, Ivana (referee) ; Matoušek, Petr (advisor)
The work focuses on the identification of mobile applications in encrypted traffic based on TLS fingerprints. The aim of the work was to create an architecture for obtaining selected attributes from TLS connection handshake, to create TLS fingerprints and their comparison. Emphasis was placed on the accuracy of individual metrics, the quality of selected attributes and on the determination of the threshold T comparison, which was ultimately set at 75 %. A total of ten attributes were selected from the TLS connection handshake, such as IP address, Cipher Suite, Server Name Indication, the size of the first ten packets and more. Accurate, substring and index comparisons were chosen to compare individual attributes. The total similarity of the two TLS fingerprints is then calculated as the weighted sum of the matches of the individual attributes. The resulting architecture allows you to compare TLS application fingerprints from the created dataset with newly created fingerprints from encrypted communication, and thus identify the applications. It also allows manual or automatic learning of new applications from the compared file, or updating of known TLS fingerprints of applications in the dataset.
|
|
|
Application for Outlier Detection
Siladi, František ; Bartík, Vladimír (referee) ; Burgetová, Ivana (advisor)
The goal of this work is to investigate methods for outlier detection and to create an application which is able to correctly identify these values using individual methods or combinations of outlier detection methods. Another goal of the application is to clearly display the results of outlier detection and then visualize them in 2D or 3D space. The work also includes experiments on selected data sets, which are adapted to outlier detection and on automatically generated sets. The application and experiments were created in Python and Qt for Python was used to create the graphical user interface.
|
|
|
Application for Statistical Analysis of ICS Communication
Chimenti, Andrea ; Bartík, Vladimír (referee) ; Burgetová, Ivana (advisor)
This work aims to present the design and implementation of an application for statistical analysis of network traffic in ICS (Industrial Control Systems) communication. In the first place, the work presents Industrial Control Systems and some of their most common protocols. The protocol IEC 104 is described in more detail. This is followed by an introduction to the basic methods of descriptive statistics, that can be used to analyze industrial communication. For this purpose, several CSV datasets, that capture fragments of industrial communication, have been used. These datasets are used to show how some of the previously described statistical methods can be used. The work then describes the implementation of an application, which allows to analyze the datasets and obtain various statistics and a visual representation of the data. The main objective of the application is to make it easier for the user to find stable characteristics that can be used for anomaly and attack detection. Finally, the benefits that the application brings are demonstrated on a set of datasets containing different types of attacks.
|
|
|
Platform for Biological Sequence Analysis Using Machine Learning
Lacko, Dávid ; Burgetová, Ivana (referee) ; Martínek, Tomáš (advisor)
Strojové učenie má veľa aktívnych odvetví a jedným z nich je charakterizácia proteínov, pretože experimentálne získavanie charakteristík je drahé a časovo náročné, a taktiež preto, že každoročne sú publikované mnohé sady údajov vhodné na trénovanie takýchto prediktorov. Jedna z nedávno vyvinutých metód, nazývaná innov'SAR, ktorá bola použitá už v niekoľkých aplikáciách proteínového inžinierstva, kombinuje Fourierovu transformáciu z čiastočnou lineárnou regresiou. Avšak, jej implementácia nie je voľne dostupná a samotná metóda nebola štatisticky overená. Cieľom tejto práce je adresovať tieto nedostatky, implementovať túto metódu v jazyku Python, rozšíriť ju a zahrnúť do ľahko použiteľnej platformy, ktorá umožní trénovanie a testovanie modelov. Taktiež bolo vykonané testovanie štatistickej významnosti za účelom overenia dopadu nájdených závislostí medzi sekvenciami a vlastnosťami proteínov. Metóda sa osvedčila ako štatisticky významná so silnými závislosťami nájdenými medzi vstupmi a výstupmi. Novo zozbierané dátové sady haloalkán dehalogenáz sa použili na vytvorenie modelov s validačným skóre Q2 = 0.54 a Q2 = 0.77, čo je takmer dvojnásobné zlepšenie oproti základným modelom. Tieto modely majú potenciál na filtrovanie väčších databáz sekvencií a vyhľadávanie proteínov s potenciálne lepšími vlastnosťami.
|
|
|
System For Analysis of Biathlon Statistics
Zeman, Ondřej ; Burgetová, Ivana (referee) ; Bartík, Vladimír (advisor)
The goal of this bachelor's thesis is to create a web application that downloads biathlon data and statistics and use the technique to get knowledge from data to get interesting and unusual information. In this work are solving descriptive and predictive mining tasks. Clustering algorithms have been used for descriptive mining tasks and for search patterns in course and shooting statistics. Prediction of race results is solved by using multiple linear regression. The application is implemented in Python. Web application is available at https://analysisofbiathlonstatistics.herokuapp.com.
|
|
|
Data Analysis of a Company Producing Medical Supplies
Kulhánková, Monika ; Bartík, Vladimír (referee) ; Burgetová, Ivana (advisor)
This bachelor's thesis deals with the analysis of the company's sales data, specifically the classification of the customer's type according to his sales data. It provides a theoretical introduction to data mining. It describes the classification process and methods for creating classifiers and presents the CRISP-DM model. This thesis describes the provided data sets, from which the relevant attributes are selected. The data are preprocessed and used in the creation and testing of classification models. The result of this thesis is a comparison of the achieved results.
|
|
|
Analysis of Outlier Detection Methods
Labaš, Dominik ; Bartík, Vladimír (referee) ; Burgetová, Ivana (advisor)
The topic of this thesis is analysis of methods for detection of outliers. Firstly, a description of outliers and various methods for their detection is provided. Then a description of selected data sets for testing of methods for detection of outliers is given. Next, an application design for the analysis of the described methods is presented. Then, technologies are presented, which provide models for described methods of detection of outliers. The implementation is then described in more detail. Subsequently, the results of experiments are presented, which represent the main part of this thesis. The results are evaluated and the individual models are compared with each other. Lastly, a method for accelerating outlier detection is demonstrated.
|
guest ::
